AIM: To identify the pre-X region in hepatitis B virus (HBV)genome and to study the relationship between the genotype and the pre-X region. To investigate the biological function of whole-X (pre-X plus X) protein, we ...AIM: To identify the pre-X region in hepatitis B virus (HBV)genome and to study the relationship between the genotype and the pre-X region. To investigate the biological function of whole-X (pre-X plus X) protein, we performed yeast two-hybrid to screen proteins in liver interacting with whole-X protein.METHODS: The pre-X region of HBV was amplified by polymerase chain reaction (PCR) method, and was cloned to pGEM Teasy vector. After the target region was sequenced, Vector 8.0 software was used to analyze the sequences. The whole-X bait plasmid was constructed by using yeast two-hybrid system 3. Yeast strain AH109 was transformed. After expression of the whole-X protein in AH109 yeast strains was proved, yeast two-hybrid screening was performed by mating AH109 with Y187 containing liver cDNA library plasmid. The mated yeast was plated on quadruple dropout medium and assayed for α-gal activity. The interaction between whole-X protein and the protein obtained from positive colonies was further confirmed by repeating yeast two-hybrid. After extracting and sequencing of plasmid from blue colonies, we carried out analysis by bioinformatics. RESULTS: After sequencing, 27 of 45 clones (60%) were found encoding the pre-X peptide. Eighteen of twenty-seven clones (66.7%) of pre-X coding sequences were found from genotype C. Five positive colonies that interacted with whole-X protein were obtained and sequenced; namely, fetuin B, UDP glycosyltransferase 1 family-polypeptide A9, mannose-P-dolichol utilization defect 1, fibrinogen-B beta polypeptide, transmembrane 4 superfamily member 4CD81 (TM4SF4).CONCLUSION: The pre-X gene exists in HBV genome.Genes of proteins interacting with whole-X protein in hepatocytes were successfully cloned. These results brought some new clues for studying the biological functions of whole-X protein.展开更多
针对现有医学影像合成技术在准确捕捉复杂解剖结构和病理状态方面存在不足,从而生成低质量且与实际情况不符的胸片问题,文中提出了一种创新性的医学潜在扩散模型Chest-Chat。基于先前研究结果改进了所提模型,引入一种多模态文本编码器Me...针对现有医学影像合成技术在准确捕捉复杂解剖结构和病理状态方面存在不足,从而生成低质量且与实际情况不符的胸片问题,文中提出了一种创新性的医学潜在扩散模型Chest-Chat。基于先前研究结果改进了所提模型,引入一种多模态文本编码器MedA-BERT(Medical Attention Strategy Pre-training of Deep Bidirectional Transformers for Language Understanding)。采用跨模态视觉-语言预训练策略构建该编码器并强化胸片影像与对应文本报告间的深刻语义联系,结合双向交叉注意力机制和对比学习显著增强了模型对医学影像报告语义的理解和处理能力。将MedA-BERT与潜在扩散模型的视觉模块相结合,使Chest-Chat能够生成具有详细解剖和病理描述的高质量胸片。在CheXpert和MIMIC-CXR(Chest X-ray)两个公开数据集上进行了广泛评估。实验结果表明,Chest-Chat的FID InceptionV3(Fréchet Inception Distance)、FID XRV和MS-SSIM(Multi-Scale Structural Similarity)分别为58.38、3.69和0.12±0.11,其表现优于现有方法。展开更多
基金Supported by the grants from the National Natural Science Foundation, No. C03011402, No. C30070690the 9.5 Research and Technique Foundation of PLA, No. 98D063+1 种基金 the Launching Foundation for Student Studying Abroad of PLA, No. 98H038 the 10.5 Youth Research and Technique Foundation of PLA, No. 01Q138andNo. 01MB135
文摘AIM: To identify the pre-X region in hepatitis B virus (HBV)genome and to study the relationship between the genotype and the pre-X region. To investigate the biological function of whole-X (pre-X plus X) protein, we performed yeast two-hybrid to screen proteins in liver interacting with whole-X protein.METHODS: The pre-X region of HBV was amplified by polymerase chain reaction (PCR) method, and was cloned to pGEM Teasy vector. After the target region was sequenced, Vector 8.0 software was used to analyze the sequences. The whole-X bait plasmid was constructed by using yeast two-hybrid system 3. Yeast strain AH109 was transformed. After expression of the whole-X protein in AH109 yeast strains was proved, yeast two-hybrid screening was performed by mating AH109 with Y187 containing liver cDNA library plasmid. The mated yeast was plated on quadruple dropout medium and assayed for α-gal activity. The interaction between whole-X protein and the protein obtained from positive colonies was further confirmed by repeating yeast two-hybrid. After extracting and sequencing of plasmid from blue colonies, we carried out analysis by bioinformatics. RESULTS: After sequencing, 27 of 45 clones (60%) were found encoding the pre-X peptide. Eighteen of twenty-seven clones (66.7%) of pre-X coding sequences were found from genotype C. Five positive colonies that interacted with whole-X protein were obtained and sequenced; namely, fetuin B, UDP glycosyltransferase 1 family-polypeptide A9, mannose-P-dolichol utilization defect 1, fibrinogen-B beta polypeptide, transmembrane 4 superfamily member 4CD81 (TM4SF4).CONCLUSION: The pre-X gene exists in HBV genome.Genes of proteins interacting with whole-X protein in hepatocytes were successfully cloned. These results brought some new clues for studying the biological functions of whole-X protein.
文摘针对现有医学影像合成技术在准确捕捉复杂解剖结构和病理状态方面存在不足,从而生成低质量且与实际情况不符的胸片问题,文中提出了一种创新性的医学潜在扩散模型Chest-Chat。基于先前研究结果改进了所提模型,引入一种多模态文本编码器MedA-BERT(Medical Attention Strategy Pre-training of Deep Bidirectional Transformers for Language Understanding)。采用跨模态视觉-语言预训练策略构建该编码器并强化胸片影像与对应文本报告间的深刻语义联系,结合双向交叉注意力机制和对比学习显著增强了模型对医学影像报告语义的理解和处理能力。将MedA-BERT与潜在扩散模型的视觉模块相结合,使Chest-Chat能够生成具有详细解剖和病理描述的高质量胸片。在CheXpert和MIMIC-CXR(Chest X-ray)两个公开数据集上进行了广泛评估。实验结果表明,Chest-Chat的FID InceptionV3(Fréchet Inception Distance)、FID XRV和MS-SSIM(Multi-Scale Structural Similarity)分别为58.38、3.69和0.12±0.11,其表现优于现有方法。
